FISHERY BULLETIN: VOL. 87, NO. 1 



ming than would larger animals entangled in a 

 similar-sized fragment (Feldkamp 1985). 



Many questions remain unanswered concerning 

 the impact of marine debris on the demographics 

 of northern fur seals. While it is virtually impossi- 

 ble to directly measure mortality arising from en- 

 tanglement in different age and sex classes, mea- 

 surements can be made of the behavioral changes 

 and energetic costs associated with entanglement, 

 the susceptibility of different age classes to en- 

 tanglement, and the effects of net size on these 

 parameters. In this study we examine the energetic 

 and behavioral costs associated with entanglement. 

 The swimming metabolic rate of three juvenile male 

 northern fur seals entangled in various-sized nets 

 was measured. Changes in behavior and average 

 daily energy expenditure during extended periods 

 of entanglement were quantified. The northern fur 

 seals' responses to floating debris, the likelihood of 

 entanglement, and their ability to free themselves 

 after entanglement were also examined to provide 

 a better understanding of the biological conse- 

 quences of net entanglement in these animals. 



MATERIALS AND METHODS 



Three newly weaned male northern fur seal pups 

 (age = 4 months, based on estimated birth date of 

 July 1; Gentry 1981) were captured on St. Paul 

 Island, AK in November 1985. They were trans- 

 ported to the Marine Laboratory, University of 

 California at Santa Cruz and placed in a large hold- 

 ing tank supplied with filtered seawater. Twice per 

 day, the animals were fed a ration of herring sup- 

 plemented with vitamins. 



These three northern fur seal pups were weighed 

 weekly. Measurements of standard length (nose to 

 tip of tail) and of girth around the head (at ears), 

 neck, and shoulder region were made at several 

 month intervals. Girths were converted to diameters 

 by assuming a circular circumference. 



Net fragments used in this study were all cut from 

 polypropylene trawl nets found on St. Paul Island, 

 AK. Each fragment had a stretched mesh size of 23 

 cm (9 in). The twine had a diameter of 3 mm (Yg in). 



Swimming Energetics 



The energetic cost of swimming, before and dur- 

 ing entanglement, was measured by placing the 

 northern fur seals in a water flume constructed in- 

 side of a circular tank, 7.6 m in diameter and 2.7 

 m deep. A wooden ring (4.9 m in diameter and 1.2 

 m in height) was placed in the tank, forming a 1.3 



m wide channel between it and the tank wall. A 

 water current was generated inside this channel 

 with two pumps. The first, a 15 hp pump, was sub- 

 merged to a depth of 48 cm and produced a flow of 

 0.75 m/s. The second, a 10 hp nonsubmersible pump, 

 was located above the tank with its intake and outlet 

 hoses fixed in the channel; this pump could generate 

 flows of 0.6 m/s. Run simultaneously, the two pumps 

 created flows of 1.1 m/s. 



The fur seals swam inside of a metabolic test sec- 

 tion (2.2 m in length, 1.1m wide, 0.9 m deep) con- 

 structed in the channel. The walls of the tank and 

 inner ring formed its sides. Front and back ends 

 were framed with wood and covered with 8 cm x 

 13 cm mesh wire screen. A sheet of plywood cov- 

 ered the top. A plastic dome (0.9 m x 0.6 m x 0.3 

 m) was set into this plywood and served as an open 

 circuit metabolic chamber. Animals in the test sec- 

 tion could only surface to breathe inside of the dome. 



To minimize turbulence, the test section was 

 located approximately 7.5 m away from the outflow 

 of the pumps, along the tank's circumference. Water 

 velocity was measured with a General Oceanics 

 Model 2035 MKIIP flow meter, accurate to ±3%. 

 All flow measurements were made in the test sec- 

 tion, 50 cm from the floor and 90 cm from the front 

 screen. 



Air was drawn through the metabolic dome at a 

 rate of 20 L/min. Oxygen content of the air was 

 measured with an Ametek oxygen analyzer cali- 

 brated using the methods of Fedak et al. (1981). A 

 computer monitored the analyzer's output each sec- 

 ond and produced a 1 min average of the percent 

 0, concentration. Oxygen consumption (VO2) was 

 calculated using equation 11 of Fedak et al. (1981). 

 Every 10 minutes, these minute readings were 

 averaged to provide single data points at each swim- 

 ming speed. All values were corrected to STPD, and 

 VO2 (in mL Oo • min "' • kg"') was converted to 

 W/kg by assuming a caloric equivalent of 20. 1 J/mL 

 O2 (Bartholomew 1977). 



Prior to the actual measurements, the three north- 

 ern fur seals were trained for several weeks to swim 

 in the flume. Training was considered complete 

 when consistent values for VO2 were obtained at 

 each speed. During experiments, 12 h fasted animals 

 were placed in the flume and allowed to rest, groom, 

 or swim at their own speed for approximately 15 

 minutes while VO2 was monitored. The first water 

 pump was then turned on and water velocity main- 

 tained at 0.75 m/s. Ten minutes were allowed for 



^Reference to trade names does not imply endorsement by the 

 National Marine Fisheries Service, NOAA, 



86 



